Intelligent Controller for a Networked Permanent Magnet Synchronous Motor

This paper is mainly concerned with the development of a control method for a permanent-magnet synchronous motor (PMSM) which runs in the conditions of a networked control system (NCS). A middleware is implemented and modifies the output of the controller based on gain scheduling. The gain scheduling is designed by using an artificial neural network, which is first trained offline to learn the relation between constant delays and the corresponding optimal gains. The search of the gain is transformed into an optimization problem and the solution is given by a Genetic Algorithm. Afterwards, considering an UDP/IP channel for the communication between the controller and the PMSM, the delays are random and varying. Data dropouts are also considered, the PMSM being controlled with the last valid command. The experiments are conducted using a DSP based drive. Advantages of the proposed method are pointed out and experimental results are presented to show the performance of the proposed method.